BACKGROUND OF THE INVENTION
Field of the Invention
[0001] The present invention generally relates to solar cells, and more specifically to
a solar cell assembly that includes a bypass diode that is positioned in a recess
on the backside of a solar cell for anti-parallel connection with the solar cell.
Description of the Related Art
[0002] Arrays which may include hundreds or thousands of solar cells bonded to a solar panel
are used to provide electrical power for a variety of applications such as spacecraft.
In most applications in which solar cells are used, size and weight considerations
are very important. A solar cell includes a flat photovoltaic wafer made from n-type
or p-type crystalline semiconductor material, such as silicon, gallium-arsenide or
germanium in or on which a thin surface layer of the opposite conductivity type is
formed. The interface between the surface layer and the main or bulk region of the
wafer defines a semiconductor junction. A current collector grid metallization is
formed on the surface layer and a metal layer is formed on the back surface of the
wafer. Illumination of the surface layer causes a liberation of charge carriers, including
electrons and holes in the region of the semiconductor junction, which migrate toward
opposite surfaces to establish a potential across the solar cell.
[0003] Solar cells are typically modeled as diodes that respond to illumination by becoming
forward biased and establishing a voltage across the cell. A silicon solar cell produces
about 0.6V, while a gallium-arsenide cell produces about 1.0V. A relatively new class
of multijunction solar cells, formed from a combination of group III and group V materials
and commonly referred to as Advanced "III-V" cells, produce a somewhat higher forward
voltage. When the solar cell is in a circuit and is not illuminated, the current flow
through the cell becomes limited, which causes the cell to become reverse biased.
If the reverse bias voltage is high enough the solar cell may break down and become
permanently damaged. Silicon cells have a typical break down voltage of approximately
60V with a minimum of approximately 35V. Gallium arsenide cells have a much lower
break down voltage, nominally 5V and as low as 1V, and hence are more susceptible
to damage. The Advanced "III-V" cells can break down at even lower voltages.
[0004] The solar cells may be connected in series strings to provide a desired voltage,
in parallel to provide a current, or in a series-parallel combination. When all of
the solar cells are illuminated, they each produce their respective voltage or current
signals which sum together to maintain the desired output. However, if one or more
of the solar cells becomes shadowed those cells become reverse biased. For example,
in a spacecraft an antenna may cast a shadow across the array.
[0005] The effect of shadowing a solar cell in a series string depends upon the specific
characteristics of the cell. If the cell has a very low reverse current, reverse biasing
the cell will effectively force the string output to zero. Conversely, if the cell
breaks down at a relatively low reverse voltage, the effect of shadowing a cell on
the string output is reduced. However, the cell can be permanently damaged.
[0006] Bypass diodes, typically silicon, are used to minimize output losses and to protect
cells when they became shadowed. Bypass diodes can be connected across single cells,
across strings of cells or across rows of parallel-connected cells. As shown in FIG.
1, the bypass diodes 10 and solar cells 12 are connected in an anti-parallel configuration
such that the bypass diode is reverse biased when the solar cells are illuminated.
Bypass diodes that have very low reverse currents are preferred to avoid reducing
current in the solar cell during normal operation, which would reduce power efficiency.
When the cell becomes shadowed, the current flow through the cell is limited, causing
the cell to become reverse biased. This causes the bypass diode to become forward
biased and conduct so that the currant in the string can continue to flow. The bypass
diode also limits the reverse bias voltage across the cell so that it does not break
down and become permanently damaged. The voltage output V
out of the string of cells is reduced by the voltage that would be produced by the non-illuminated
cells and the voltage drop across the bypass diode.
[0007] The "Solar Cell Array Design Handbook," by H.S. Rauschenbach,
Litton Education Publishing, Inc., pp. 300-302, 1980 discloses three types of bypass diodes: conventional rectifier
diodes, rectifier wafers and integral diodes. The conventional rectifier diodes are
wire bonded to the solar panel adjacent the solar cells. In strings of silicon cells,
one bypass diode per every seventy cells is sufficient to limit the maximum reverse
voltage to be less than 35V so that the cells do not break down. The additional space
on the solar panel, weight, time and cost of installing one diode per seventy cells
is not significant. However, in strings of gallium-arsenide cells, one bypass diode
per cell would be required to ensure protection. Instead, typically one diode per
five cells is used and the cells are screened to remove those cells that break down
at less than 5V. This typically results in a loss of between 10% and 30% of the cells.
Furthermore, wire bonding the diodes to the surface of the solar panel generally increases
the overall size, and thus weight, of the panel by approximately 15%.
[0008] In Rauschenbach, rectifier wafers are used to protect solar cells in densely packed
arrays in order to conserve space. The rectifier wafers are placed underneath the
solar cells but are not directly attached to the cells. Solar cells are bonded to
the solar panel along an adhesive bond line that is typically 0.004 inches thick.
The placement of the wafers between the solar cells and the panel necessitates the
use of a thicker bond line. This increases the cost of manufacture, substantially
increases the weight of the solar panel and may weaken the mechanical connection between
the solar cells and panel.
[0009] Rauschenbach discloses that a significant improvement over the flat wafers can potentially
be achieved by using solar cells with integral diodes. An integral diode is formed
by fabricating the diode into the solar cell itself. This can be accomplished for
silicon solar cells but is very difficult for gallium-arsenide cells. Furthermore,
integral diodes are characterized by a significant reverse current that tends to degrade
solar cell performance.
SUMMARY OF THE INVENTION
[0010] The present invention seeks to provide a compact lightweight solar cell-bypass diode
assembly that has high mechanical strength and low reverse current.
[0011] This is accomplished by forming at least one recess in the back (non-illuminated)
side of the solar cell and placing at least one discrete low-profile bypass diode
in respective recesses so that each bypass diode is approximately flush with the back
side of the solar cell. Each bypass diode is bonded to the solar cell for anti-parallel
connection across the solar cell.
[0012] In the preferred embodiment, the back side of the solar cell is formed with a honeycomb
pattern of recesses to reduce the weight of the solar cell while maintaining mechanical
strength. The recesses that receive bypass diodes are preferably modified to have
a rectangular shape that better accommodates the bypass diodes.
[0013] For a better understanding of the invention, and to show how the same may be carried
into effect, reference will now be made, by way of example, to the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014]
FIG. 1, described above, is a schematic diagram of a prior series string of solar
cells connected in an anti-parallel configuration with a bypass diode;
FIG 2 is a simplified sectional view of a plurality of solar cells and respective
planar isolated bypass diodes that are formed in recesses on the back surfaces of
the solar cells in accordance with the present invention;
FIG. 3 is a plan view of the solar cells and bypass diodes illustrating their interconnections;
FIG. 4 is a schematic diagram of a series string of solar cells having at least one
bypass diode connected across each solar cell;
FIG. 5 is a plan view of the back surface of a preferred solar cell-bypass diode assembly;
FIG. 6 is a sectional view of a plurality of solar cell-bypass diodes that are connected
in a series string and bonded to a solar panel; and
FIG. 7 is a simplified sectional view of the solar cells in FIG. 2 with respective
mesa isolated bypass diodes.
DETAILED DESCRIPTION OF THE INVENTION
[0015] The present invention provides a solar cell-bypass diode assembly in which a discrete
bypass diode is positioned in a recess formed in the back surface of the solar cell
and is bonded to the solar cell. The solar cell-bypass diode assembly requires less
space than conventional rectifier diodes, provides a better mechanical bond to the
solar panel than the rectifier wafers, and has lower reverse current than the integral
diode. Redundant bypass protection can be provided by forming more than one recess
on the back side of each solar cell, placing a bypass diode in each recess, and connecting
each of them in anti-parallel with the solar cell. Furthermore, the electrical connections
between series connected solar cells and the bypass diode electrical connections can
be made simultaneously, thereby improving the electrical and mechanical quality of
the connection.
[0016] FIG. 2 is a simplified cross-sectional view of a series string of solar cell assemblies
14 that each include a solar cell 16 and at least one discrete bypass diode 18. As
shown, solar cell 16 is an n-type germanium (Ge) wafer 20, suitably having a thickness
of 0.007 inches (178µm) and a cross-sectional area of 2.54 cm x 5.08 cm. Alternately,
the wafer could be p-type Ge, p or n-type silicon, or a III-V multi-junction material.
A thin, suitably 10µm or less, p-type layer 22 is formed on the solar cell's front
surface 24 by means of a metallo organic chemical vapor deposition (MOCVD) procedure
or a diffusion procedure. Conductive grid lines 26 formed on the front surface 24
of solar cell 16 collect current and deliver it to a collector bar 28 that is connected
to a flexible interconnect tab 30. The tab 30 is preferably formed from Kovar in a
U-shape and provides the solar cell's p-contact (anode).
[0017] The solar cell 16 has at least one recess 32 formed in its back surface 34. The recess
32 is preferably formed by etching the back surface 34 of the solar cell through a
mask either by microblasting or conventional photolithography. The recess 32 has a
depth of approximately 0.004 inches (100µm) and a rectangular cross sectional area
of approximately 0.44 cm x 0.44 cm. A metalization layer 36 is preferably formed over
the solar cell's back surface 34 and recess 32 after the recess has been etched, and
provides the solar cell's n-contact (cathode).
[0018] Bypass diode 18 is preferably a low-profile discrete diode, suitably 0.003-0.005
inches (76.2µm-127µm) thick with a cross-sectional area of 14.4mm
2 (3.8 mm on a side), that is fabricated using conventional semiconductor processing
techniques. The planar isolated bypass diode 18 is bonded to the solar cell 16 inside
recess 32 by a silicone adhesive layer 38 that is approximately 0.001 inches (25µm)
thick. The low-profile diode's front surface 40 is preferably flush with the back
surface of the solar cell, but may extend 0.001-0.002 inches (25-50µm) past the back
surface. The standard bond line between the solar cell and a solar panel is 0.004
inches (100µm). Thus, the solar cell-bypass diode assembly can be bonded to the solar
panel using the standard bond line. This simplifies subsequent manufacturing and improves
the mechanical bond between the solar cell and panel.
[0019] The bypass diode 18 shown in FIG. 2 is a p-type silicon planar isolated diode, formed
by diffusing an n-type region 42 into a p-type wafer 44 to define a pn junction at
their interface. The bypass diode is formed from a semiconductor material, such as
silicon, that has a turn on voltage that is less than the reverse break down voltage
of the solar cell. The diode preferably has a very low, ideally zero, reverse current.
The bypass diode's anode and cathode are provided by a p-contact 46 formed over the
p-type wafer 44 and an n-contact 48 formed over the n-type region 42.
[0020] As shown in FIGs. 2 and 3, the solar cells are connected in series by bending the
flexible interconnect tabs 30 (p-contacts) on the front surfaces 24 of the solar cells
and contacting them to the metal layers 36 (n-contacts) on the back surfaces 34 of
the adjacent solar cells. The bypass diodes are connected in anti-parallel with their
associated solar cells by placing conductive interconnects 50 across the n-contact
48 on one assembly and the metal layer 36 and p-contact 46 on the next assembly. The
interconnects 50 are suitably silver foil having a thickness of 0.0005 inches (12.7µm).
The adhesive layer 38 also forms an insulator between the interconnect 50 and the
metal layer 36 at the bypass diode's n-contact 48. The tabs 30 and interconnects 50
are simultaneously welded to the metal layer 36 and diode contacts 46 and 48. This
improves the quality of the weld and reduces manufacturing costs by forming the two
connections simultaneously.
[0021] As shown in FIG. 4, the solar cell-bypass diode assembly and interconnection scheme
illustrated in FIGs. 2 and 3 results in at least one bypass diode 18 being connected
in anti-parallel with each solar cell 16. The solar cells may be formed with multiple
recesses so that more than one bypass diode can be connected in anti-parallel across
the solar cell. Thus, if one bypass diode is defective or is damaged the other diodes
will provide redundancy. The redundant diodes are shown with dashed lines because
they may or may not be included. This is the preferred configuration because 1) it
provides high break down protection for the solar cells by limiting the reverse voltage
that can occur across each cell and by providing redundant bypass diodes, and 2) it
is simple to connect the bypass diode across its associated solar cell. Alternately,
groups of n solar cells in a string could be provided with a single bypass diode as
shown in FIG. 1. This would provide less protection and would be more difficult to
connect.
[0022] As shown in FIG. 5, the back surface 34 of solar cell 16 is preferably formed with
a generally regular pattern of recesses. U.S. Patent Application, "Thin Semiconductor
Device and Method of Fabrication," Serial No. 08/415,221, filed March 28, 1999 filed
as a continuation of Serial No. 08/168,112 filed December 14, 1994, which is assigned
to Hughes Aircraft Company, the assignee of the present invention, discloses a solar
cell with a recess pattern formed on its back surface. The recess pattern is formed
to reduce the weight of the solar cell while retaining its physical strength. The
recesses 32 are preferably formed by microblasting the back surface 34 through a mask
to etch the n-type layer. A honeycomb pattern provides a high degree of mechanical
strength for the solar cell. Microblasting entails mixing a fine aluminum oxide powder
with a stream of compressed gas to etch the back surface 34. Microblasting is typically
used to thin and roughen silicon wafers, but can also be used to etch the recesses
in accordance with the invention. In a preferred embodiment of the invention, the
recesses 32 which receive bypass diodes 18 have a rectangular rather than a hexagonal
shape for a better fit of the bypass diodes 18. As shown, the solar cell includes
two bypass diodes, which provide redundant protection for the cell.
[0023] FIG. 6 is a sectional view of the solar cell-bypass diode assemblies 14 mounted on
a solar panel 52. The back surface 34 of solar cell 16 and the front surface 40 of
bypass diode 18 are bonded to the solar panel 52 with a silicon adhesive. Because
the bypass diode has a low profile that extends at most approximately 0.002 inches
(50.8µ) past the solar cell's back surface, conventional manufacturing processes can
be used to provide a bond line 54 having a standard thickness of 0.004 inches (101.6µ).
This improves the quality of the bond and reduces manufacturing costs.
[0024] FIG. 7 shows an alternate embodiment of the invention that uses a mesa isolated bypass
diode 56 in place of the planar isolation diode 18. The mesa bypass diode 56, in this
example, includes a p-type silicon wafer 58 that has a thin n-type layer grown or
diffused on its front surface. An n-contact 60 is formed over the n-type layer and
a p-contact 62 is formed on the other side of the p-type wafer. Instead of forming
separate mechanical and electrical connections between the bypass diode and the solar
cell, the p-contact 62 is soldered directly to the metal layer 36 on the back surface
34 of the solar cell 36. This provides the mechanical connection between the bypass
diode and solar cell and the electrical connection between the bypass diode's p-contact
and the solar cell's n-contact. An interconnect 64 is still used to connect the bypass
diode's n-contact 60 to the solar cell's p-contact 36.
[0025] While several illustrative embodiments of the invention have been shown and described,
numerous variations and alternate embodiment will occur to those skilled in the art.
Such variations and alternate embodiments are contemplated, and can be made without
departing from the spirit and scope of the invention as defined in the appended claims.
1. A solar cell assembly (14), comprising:
a solar cell (16) having a front surface (24), a back surface (34), a first contact
(28) on the front surface (24), and a second contact (36) on the back surface (34),
said solar cell (16) producing a voltage signal between said first and second contacts
(28, 36) in response to illumination of its front surface (24); and
a bypass diode (18; 56) that is bonded to the solar cell (16), said bypass diode
(18; 56) having third and fourth contacts (46, 48; 62, 60) connected to said first
and second contacts (28, 36), respectively, in an anti-parallel configuration so that
said bypass diode (18; 56) is reverse bias when said solar cell (16) is illuminated
and is otherwise forward biased to limit the reverse bias voltage across the solar
cell (16) and prevent it from breaking down,
characterized by
a recess (32) being formed in the back surface (34) of the solar cell (16), and
said bypass diode (18; 56) being positioned in said recess (32).
2. The solar cell assembly of claim 1, characterized in that said bias diode (18) is
a planar isolated diode (18) that has a back surface bonded to the solar cell (16)
and a front surface (40) on which said third and fourth contacts (46, 48) are positioned.
3. The solar cell assembly of claim 2, characterized in that said second contact (36)
covers the back surface (34) of the solar cell (16) and said bypass diode (18) is
adhesively bonded to the second contact (36).
4. The solar cell assembly of claim 1, characterized in that said bypass diode (56) is
a mesa isolated diode (56) that has back and front surfaces on which said third and
fourth contacts (62, 60) are positioned, respectively, said third contact (62) being
bonded to the solar cell's second contact (36) to provide a physical and electrical
interconnection.
5. The solar cell assembly of any of claims 1 - 4, characterized in that said bypass
diode (18; 56) has a front surface (40) that is approximately flush with the back
surface (34) of said solar cell (16).
6. The solar cell assembly of any of claims 1, 2 or 4, characterized in that the back
surface (34) of the solar cell (16) has a plurality of said recesses (32), said bypass
diode (18; 56) being positioned in one of said recesses (32).
7. The solar cell assembly of claim 6, characterized in that said recess (32) that receives
said bypass diode (18; 56) has one shape and the remaining recesses have a different
shape.
8. The solar cell assembly of claim 7, characterized in that said remaining recesses
are hexagonal and are arranged in a honeycomb pattern.
9. A method for forming a solar cell assembly (14) comprising the steps of:
providing a solar cell (16) having a front surface (24), a back surface (34), a
first contact (28) on the front surface (24), a second contact (36) on the back surface
(34), said solar cell (16) producing a voltage signal between said first and second
contacts (28, 36) in response to illumination of its front surface (24); and
providing a bypass diode (18; 56) having third and fourth contacts (46, 48; 62,
60) connected to said first and second contacts (28, 36), respectively,
characterized by
forming a recess (32) in said back surface (34) of said solar cell (16), and
positioning said bypass diode (18; 56) in said recess (32) and bonding it to said
solar cell (16).
10. The method of claim 9, characterized in that the step of providing said solar cell
(16) comprises:
providing a photovoltaic wafer (20) of a predetermined conductivity type, said
wafer (20) having front and back surfaces;
forming a thin layer (22) of opposite conductivity on the front surface of the
photovoltaic wafer (20) to form a semiconductor junction;
forming said first contact (28) over said thin layer (22);
removing a portion of said photovoltaic wafer (20) to form said recess (32) in
its back surface (34); and
forming said second contact (36) over the wafer's back surface (34).